Intonational Marking of Focus in L2 Italian and French and the Markedness Differential Hypothesis

Résumé En

L2 prosody learning models typically consider the degree of similarity between a learner's L1 and L2. However, other factors, such as markedness and learning direction, have not been systematically examined in this context. The limited application of the Markedness Differential Hypothesis (MDH) [1] in prosody may stem from the complexity of defining 'markedness' in relation to sentence prosody. According to pioneering works [2, 3], a markedness scale can also be conceptualized for prosody: in this framework, structurally determined prosody is considered less marked than pragmatically determined prosody (see Fig. 1). To establish a language's position on the markedness scale, it is necessary to observe its accentuation, intonation, and discourse-level prominence marking systems. French is often characterized by high variability in the placement of prosodic prominence, due to the 'syncretism' between these features [4, 5, 6]. Italian, in contrast, exhibits a more regular and structurally determined prosodic and accentual system [7, 8, 9]. Based on these observations, Italian is classified as a language with structurally determined sentence prosody, while French occupies an intermediate position between structurally and pragmatically determined systems [2]. For the purposes of this study, this distinction implies that French is more marked than Italian. Following the MDH, it is predicted that francophone learners of L2 Italian will exhibit greater proficiency in implementing focus-marking prosody, compared to italophone learners of L2 French. This prediction aligns with the hypothesis that markedness influences L2 acquisition asymmetrically, making structurally simpler systems easier to acquire. Practically, this means we expect a higher frequency of successfully realized focus-marking prosodic contours in the L2 Italian group, while the L2 French group is anticipated to exhibit fewer instances of focus-related intonational marking.To test these hypotheses, we conducted a study involving 15 francophone learners of L2 Italian, 15 francophone learners of L2 French, 15 native monolingual Italian speakers, and 15 native monolingual French speakers. Data were elicited using a protocol adapted from Gabriel [10], which targeted focus subjects under broad, identification, and correction focus conditions within SVO utterances. Each participant produced six target utterances, resulting in a dataset of 360 tokens covering all focus types. Prosodic patterns were analyzed using the Polytonia algorithm [11], ensuring neutral, language-independent annotation.The results from native speakers revealed that broad focus subjects were realized without prosodic prominence in both Italian and French (see Fig. 2). Under narrow focus conditions, Italian speakers marked subjects prosodically in 50% of cases, regardless of focus subtype (identification or correction). In contrast, French speakers employed prominent contours in 20% of identification and 40% of correction contexts, showing significant functional differentiation. French speakers also exhibited great variability in f0 peak alignment, consistent with the characterization of French prosody as more variable and pragmatically driven [4, 5, 6]. Among L2 speakers, French learners of Italian approximated native Italian focus-marking patterns, using prominent f0 contours in 39% of identification and 60% of correction contexts. Conversely, Italian learners of French struggled with prosodic prominence, with rates not exceeding 21% in either narrow-focus condition. Unlike their French counterparts, they failed to express functional differentiation between identification and correction focus. Notably, Italian learners of French also exhibited an overall underuse of prosodic marking for focus constituents. This pattern aligns with our initial hypothesis, rooted in MDH: Italian prosody is more accessible for L2 learners, while French's system poses greater challenges. The consistent directionality of the asymmetry suggests that markedness plays a key role. Our findings highlight the importance of markedness in L2 acquisition and in designing effective teaching strategies for languages with divergent prosodic systems.